Spontaneous abortion (SAB) is the most common adverse outcome of pregnancy, and violence may be the most prevalent reproductive risk factor among urban, low-income women. The specific aims of this application are to : 1) characterize the prevalence and type of violence experienced by women early in pregnancy; 2) evaluate whether violence during pregnancy is an important independent risk factor for SAB; 3) evaluate whether the frequency, nature, or timing of violence affects the risk of SAB; and 4) generate hypotheses regarding prevention and intervention programs for urban, low-income pregnant women. Women 14-40 years of age from a defined inner-city community who attend the Hospital of the University of Pennsylvania's Emergency Department (ED) will be screened for pregnancy. Cases will include women who present with SAB or who experience a SAB during the follow-up period (N=465). Controls will include all women with intrauterine singleton pregnancies that continue through 20 weeks gestational age. Standardized baseline data collection will include a structured in-person interview, laboratory analyses of hair and urine for cotinine, alcohol, and drugs of abuse. Chromosomal analyses will be conducted on products of conception for approximately 120 cases who abort at HUP. Follow-up telephone interviews will take place at 16 and 20 weeks gestation. Pregnancy outcome status will also be ascertained through ongoing review of medical records, pathology logs, and birth certificates. Initial analyses will be exploratory and descriptive, characterizing cases and controls by current or past violence, age, substance abuse, social isolation. The primary analysis will be logistic regression, with odds ratios and 95% confidence intervals. The investigators will explore whether the effect of violence on spontaneous abortion varies with chromosomal status of the fetus as well as according to the timing, frequency and nature of violent incidents. Finally, they will characterize the patterns of use of health and social service agencies by abused women early in pregnancy. The investigators state that the significance of this study lies in its capacity to provide basic epidemiological data regarding the nature and circumstances of violence among pregnancy inner-city women, evaluate whether violence is an independent predictor of spontaneous abortion, and ultimately contribute information important to designing effective treatment and prevention interventions in urban settings.